Ed Beeching
Research Scientist at Hugging Face
Lyon, Auvergne-Rhône-Alpes, France
Summary
🤩
Rockstar🎓
Top SchoolEd Beeching is a Research Scientist at Hugging Face with a decade of experience bridging deep reinforcement learning, embodied AI, and practical ML engineering. He holds a PhD in Deep Reinforcement Learning from INSA Lyon/INRIA and has applied structured memory and planning approaches to navigation in large 3D environments, including work at Ubisoft LaForge. His background spans physics and six years in geophysics signal and image processing, giving him strong grounding in real-world, noisy data and scalable engineering. At Hugging Face he contributes to RLHF, LLM tool use, and performance-focused tooling—having improved RL training pipelines and memory usage in trl and implemented model evaluation and hub integration for open-r1. He is an active open-source contributor (e.g., godot_rl_agents and sample-factory) who builds practical agent environments and robust action-distribution solutions that bridge research and game-engine deployment. Colleagues rely on him for combining rigorous research with production-minded implementations that generalize beyond lab settings.
10 years of coding experience
1 year of employment as a software developer
Master's degree, Machine Learning and Data Mining, Master's degree, Machine Learning and Data Mining at Université Jean Monnet Saint-EtienneBeechen Cliff SchoolDoctor of Philosophy - PhD, Deep Reinforcement Learning, Doctor of Philosophy - PhD, Deep Reinforcement Learning at INSA Lyon - Institut National des Sciences Appliquées de LyonMaster's degree, Artificial Intelligence, Master's degree, Artificial Intelligence at KU LeuvenBachelor of Science (BS), Physics, 2:1, Bachelor of Science (BS), Physics, 2:1 at University of Surrey
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (22)
transformers10
pytorch10
godot10
python10
evaluation10
huggingface-hub10
machine-learning10
reinforcement-learning10
gdscript10
eval10
huggingface10
deep-learning10
gpt10
testing9
pytest9
Programming languages (7)
Github contributions (5)
edbeeching/godot_rl_agents
Jul 2021 - Jan 2023
An Open Source package that allows video game creators, AI researchers and hobbyists the opportunity to learn complex behaviors for their Non Player Characters or agents
Role in this project:
Back-end Developer
Back-end DeveloperContributions:6 releases, 78 reviews, 330 commits in 1 year 7 months
Contributions summary:Ed implemented core functionalities for the Godot RL Agents package. They created the initial BallChase game environment, including scene and project file configurations, as well as example environments. They also established communication protocols and action-reward mechanisms between a Python script and the Godot game engine via TCP, enabling the transfer of actions and observations. The user further updated the Jumper and FlyBy environments to use new sensors and the latest API of the Godot Engine.
aireinforcement-learningsimulationgodot
huggingface/open-r1
Jan 2025 - May 2025
Fully open reproduction of DeepSeek-R1
Role in this project:ML Engineer
ML EngineerContributions:91 reviews, 64 PRs, 118 pushes in 4 months
Contributions summary:Ed's primary contribution focused on integrating and evaluating machine learning models within the Hugging Face ecosystem. They added evaluation scripts for the AIME24 and MATH-500 datasets, indicating a focus on model performance analysis. The user also implemented automatic evaluation callbacks and integrated the models with the Hugging Face Hub, including asynchronous model pushing and managing revisions. Furthermore, the user contributed to the GrPO training arguments, suggesting involvement in training and model optimization.
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.